Introduction
Despite continuing technological advances in cardiovascular medicine, the history and physical examination remain vital to establish an accurate diagnosis of chronic coronary artery disease (CAD). Whereas many patients may describe typical angina, providers cannot rely on these symptoms alone to diagnose ischemia because silent (asymptomatic) ischemia remains a common presentation, estimated to affect almost half of patients with CAD. Conversely, some patients may endorse symptoms that closely mimic typical angina in the absence of significant CAD. Thus appropriate integration of patient symptoms, demographics, clinical characteristics, and examination findings remains essential for the clinician to accurately determine the likelihood or classification of CAD and to assess comorbidities and sequelae.
Positive interactions during the interview and examination can also lay the foundation for establishing a healthy doctor-patient relationship, critical to allowing the patient to freely share his/her experiences, goals, and preferences with the provider. Formation of such trust may also increase the likelihood of adherence to the treatment plan. This is of particular importance in the treatment of chronic coronary disease, which frequently relies on a multitude of medications and therapeutic lifestyle changes to improve symptoms and survival. Thus the following discussion of history-taking and physical examination may have important diagnostic and therapeutic implications for chronic CAD.
History
Typical Angina Pectoris
Typical anginal pain is characterized by its location, quality, duration, and exacerbating/alleviating factors. Angina can be retrosternal or diffuse in the chest, but it can also be felt in regions corresponding to the C7 through T4 dermatomes (eg, neck, jaw, and arms) because the sympathetic afferent nerves that signal myocardial ischemia also innervate these territories. Typical angina may be pressure-like, squeezing, or heavy in quality. It is usually not sharp or stabbing, pleuritic, or positional. Although the classic Levine sign (in which the patient places his/her fist over the chest to describe anginal pain) is often discussed, in a contemporary population this gesture had a low sensitivity (6%) for the detection of angina in patients admitted with chest discomfort. Angina typically lasts for minutes rather than seconds in duration, and it may be exacerbated by physical exertion or mental/emotional stress and alleviated by rest and/or nitroglycerin.
Atypical and Nonanginal Symptoms
Symptoms that do not fit the above classic features of typical angina are often referred to as atypical. Three categories of typical, atypical, and nonanginal symptoms are frequently used for clinical simplification ( Table 7.1 ). However, it is important to note that the presence of typical angina alone does not confirm the presence of CAD; conversely, atypical or nonanginal features alone do not exclude the possibility of CAD. Nonetheless, some features may increase or decrease the likelihood of chest pain being anginal in origin. Symptoms and exam features with significant positive and negative likelihood ratios for angina are shown in Fig. 7.1 . Calculation of the posttest probability of CAD with likelihood ratios (LRs) is done as follows. First, estimate the pretest probability (P pre ) of CAD and convert the probability to odds (O pre ) with O pre = P pre /(1 − P pre ). Second, multiply that pretest odds (O pre ) of CAD by the LR to determine the posttest odds (O post = LR × O pre ). Finally, convert the posttest odds back to a probability, P post = O post /(1 + O post ).
| Typical Angina | Retrosternal chest discomfort Increased with exertion or emotional stress Relief with rest or nitroglycerin |
| Atypical Angina | Exhibits 2 of the above features |
| Noncardiac Chest Pain | Exhibits 0 or 1 of the above features |
The LR with the lowest point estimate, making angina least likely, is pleuritic-type chest pain, followed by a positional component, pain being sharp/stabbing in quality, or reproducible with palpation. Angina is slightly less likely when pain is located under the breast or not associated with exertion. Although features such as association with exertion, radiation to the left arm, associated diaphoresis, nausea/vomiting, and pressure-like quality are classic/typical features of angina, these features did not lead to a large change in the probability of angina. Features with the highest LRs, which would theoretically most increase the probability of angina, were observed to be radiation to the right arm or both arms. Of note, the confidence intervals were wide for these latter two features, indicative of imprecision in the estimates of the LRs.
There are also other features of chest pain that have traditionally been accepted as useful in the differentiation of anginal versus nonanginal chest pain (eg, relief with nitroglycerin or gastrointestinal [GI] cocktails), but they may not have significant predictive value. In an analysis of 459 patients with chest pain at an urban teaching hospital, nitroglycerin relieved chest pain in 35% of patients who were found to have CAD as a cause of chest pain versus 41% of patients without CAD ( p > .20). Similarly, relief of chest pain with a GI cocktail (often consisting of viscous lidocaine, an antacid, with/without other components) has also been documented to poorly differentiate symptoms of myocardial ischemia. Some patients may not have chest pain during periods of ischemia, but instead only experience dyspnea or jaw, neck, or arm pain. These symptoms without chest pain are often called anginal equivalents . Within each individual, specific anginal equivalents (ie, dyspnea, jaw pain, etc.) may characteristically recur with each subsequent episode of ischemia, and thus providers should inquire about similarities/differences to prior known anginal/ischemic episodes.
Gender Differences in Presentation
Historically, women have often been underrepresented in major clinical trials of CAD ; thus, characterization of CAD presentation in women has largely been derived from smaller analyses. However, compared with men, women have been reported to have higher prevalence of atypical chest pain features, including pain during rest, sleep, or mental stress, and associated neck/shoulder pain, nausea, fatigue, and dyspnea with ischemia. The Women’s Ischemic Syndrome Evaluation (WISE) study reported that as many as 65% of women with CAD may not exhibit “typical” anginal symptoms. Some of these differences may be related to greater prevalence of coronary microvascular dysfunction, vasospasm, or heightened pain perception in women. Further studies have suggested that gender differences in language or descriptors used may account for some of the reported differences in presentation, in that women may be more likely to describe their symptoms with other terms such as discomfort , pressing , aching , or shortness of breath when compared with men.
Despite these reported gender differences, several studies have found that the presenting symptoms of CAD may be more similar than different between men and women. In a meta-analysis of 74 studies involving more than 20,000 patients, women reported angina at a similar or slightly higher frequency than men did. Even among many studies reporting gender differences in angina presentation, the most common descriptors used by both genders tended to be typical features such as “chest pain,” “pressure,” and “tightness.”
Fitness and Functional Capacity
Beyond inquiring about symptoms, assessing functional capacity during the history is important for assessing risk and prognosis. Multiple studies have confirmed a graded and inverse association between fitness level and mortality in patients with cardiovascular disease, independent of other risk factors. For example, in an analysis from the Duke database, patients who were able to exercise greater than 10 metabolic equivalents (METs) on a Bruce protocol without ischemia had 95% survival at 4 years; whereas those who could not achieve 4 METs had only 59% survival at 2 years. In an analysis from the Veterans Affairs database, patients were divided into quintiles of exercise capacity. Patients in the lowest quintile of fitness (achieving <5 METs) had fourfold higher adjusted relative risk for mortality at 6 years compared with those in the highest quintile (>10.7 METs). Furthermore, in an analysis of 9852 patients with known CAD from the Henry Ford Exercise Testing Project, each 1-MET increase in exercise capacity was associated with approximately 13% lower adjusted risk of mortality over the median follow-up of 11 years. Furthermore, patients with similar exercise capacity were found to have similar risk for mortality regardless of baseline revascularization status.
Estimating the Probability of Coronary Artery Disease
Relying on the description of symptoms alone is insufficient to accurately diagnose CAD. Whereas a wide array of diagnostic testing modalities, including stress electrocardiography, echocardiography, myocardial perfusion imaging, magnetic resonance imaging, coronary computed tomography, and cardiac catheterization, are available, the indiscriminate application of these testing modalities can lead to potential misclassification of patients (as incorrectly having or not having CAD) due to imperfect test sensitivities/specificities. Moreover, mounting costs and risks for complications/adverse side effects of cardiac testing also remain important considerations. Therefore, prior to deciding on whether a diagnostic test should be used and/or selecting the most appropriate modality, providers must develop an estimate of the probability of CAD for each patient undergoing evaluation.
Categorizing symptoms into typical, atypical, and nonanginal pain can increase or decrease the likelihood of CAD; however, clinicians cannot accurately estimate the probability of CAD without first considering those symptoms in the context of the patient’s age and gender. This concept was first illustrated by Diamond and Forrester in 1979 and was validated by similar findings from the Coronary Artery Surgery Study (CASS) trial. The importance of age and gender in estimating the probability of CAD in a more contemporary multinational cohort is shown in Table 7.2 . In patients presenting with typical anginal symptoms, the probability of significant CAD (>50% stenosis) can range from 28% to 93%, depending on age/gender. Likewise, the probability of CAD with nonanginal pain can range from 5% to 65%. Accordingly, many patients with nonanginal symptoms may still have higher probability for CAD than others with typical angina. For example, an 80-year-old man with nonanginal symptoms would have a 65% probability of CAD versus a 35-year-old woman with typical angina who would have a 28% probability of CAD (see Table 7.2 ).
| Age (years) | Nonanginal Pain (%) | Atypical Angina (%) | Typical Angina (%) | |||
|---|---|---|---|---|---|---|
| Women | Men | Women | Men | Women | Men | |
| 30–39 | 5 | 18 | 10 | 29 | 28 | 59 |
| 40–49 | 8 | 25 | 14 | 38 | 37 | 69 |
| 50–59 | 12 | 34 | 20 | 49 | 47 | 77 |
| 60–69 | 17 | 44 | 28 | 59 | 58 | 84 |
| 70–79 | 24 | 54 | 37 | 69 | 68 | 89 |
| >80 | 32 | 65 | 47 | 78 | 76 | 93 |
Incorporating comorbidities with age, gender, and symptoms can further improve determination of CAD probability. This was illustrated by investigators using the Duke Database of Cardiovascular Disease ( Table 7.3 ). For example, using Diamond–Forrester type classifications based on age/gender/symptoms alone, a 35-year-old man with typical angina would have a 59% probability of CAD; however, the likelihood of CAD would differ significantly between a healthy 35-year-old man with no cardiovascular risk factors and a 35-year-old man with diabetes, hyperlipidemia, and tobacco abuse. In the 35-year-old man with no risk factors, the probability of significant CAD (≥70% stenosis in a coronary artery) would be 30% versus 88% in the man with multiple risk factors. If patients had abnormal electrocardiograms at rest with significant ST- or T-wave changes or Q waves, probabilities of CAD would be higher.
| Age (years) | Nonanginal Pain (%) | Atypical Angina (%) | Typical Angina (%) | |||
|---|---|---|---|---|---|---|
| Women | Men | Women | Men | Women | Men | |
| 35 | 1–19 | 3–35 | 2–39 | 8–59 | 10–78 | 30–88 |
| 45 | 2–22 | 9–47 | 5–43 | 21–70 | 20–79 | 51–92 |
| 55 | 4–21 | 23–59 | 10–47 | 45–79 | 38–82 | 80–95 |
| 65 | 9–29 | 49–69 | 20–51 | 71–86 | 56–84 | 93–97 |
The Diamond-Forrester-type classifications and estimations from the Duke database can be of great utility in determining more precise probabilities of CAD; however, several limitations should be noted. These predictive models were developed from patients referred to university hospital settings; therefore, they may overestimate the probability of CAD for lower-risk primary care and/or community medical center populations. Additionally, these models may be less precise and overestimate probabilities of CAD in women when compared with the much lower prevalence of CAD observed in women in the WISE study. Furthermore, these probabilities were derived from eras (the 1970s and 1980s) when risk factor burdens significantly differed from contemporary populations who have significantly less tobacco use, higher prevalence of obesity and diabetes, and higher prevalence of CAD in younger individuals. Finally, these reported probabilities tend to overestimate the prevalence of CAD across all age groups and genders when compared with studies that assess the prevalence of CAD by coronary computed tomographic angiography (CTA). This may be due to probability estimates traditionally being derived from patients undergoing clinically indicated evaluation with invasive cardiac catheterization (often after abnormal stress test results) who are more likely to have disease versus low- to intermediate-risk patients preferentially referred for noninvasive testing.
Assessment and Classification of Angina
Assessment and classification of angina, including chronicity, severity, and burden of disease, are vital to determining the most appropriate treatment strategy. For example, one must ensure that an acute coronary syndrome is not present prior to instituting a treatment plan solely geared toward chronic stable CAD. Additionally, adequate assessment of the burden of symptoms and impairment in quality of life is necessary to tailor therapies to achieve adequate relief of symptoms.
Unstable angina is defined as angina that is new in onset, increasing in frequency, severity, or duration, or occurs at rest. A classification system of unstable angina was devised by Braunwald ; however, chronic stable angina is most often categorized by the Canadian Cardiovascular Society (CCS) classification system ( Table 7.4 ), ranging from class 1 angina occurring only with strenuous, rapid, or prolonged exertion to class 4 angina with inability to do any activity without symptoms and/or angina at rest.
| CCS Class | Description (Summary) |
|---|---|
| I | Angina with strenuous, rapid, or prolonged exertion |
| II | Slight limitation of ordinary activity |
| III | Marked limitation of ordinary activity, such as walking 1 to 2 blocks or climbing 1 flight of stairs |
| IV | Angina at rest |
As patients are often concerned with optimizing quality of life beyond longevity alone, further assessment of symptom-burden and health-related quality of life can be beneficial to understand the impact of CAD on patients’ lives and to assess response to therapies. Clinicians can use semiobjective quantitative measures such as how often the patient requires sublingual nitroglycerin and/or approximate walking distance achieved (eg, two blocks, 0.5 miles) before stopping due to angina. General health-related quality of life may be measured by survey instruments such as the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36) ; whereas the Seattle Angina Questionnaire (SAQ) is an example of a disease-specific quality of life instrument for CAD.
Historically, such surveys have been primarily used in research settings; however, with the advent of abbreviated survey versions and a growing emphasis on patient-centered outcomes in clinical medicine, there is potential for expanded use of such measures in clinical practice to supplement traditional history-taking for symptom assessment. The shortened version of the SAQ is a seven-question survey that can be completed at or prior to an office visit and may potentially improve efficiency of care. With the same questions asked the same way every time, the clinician may have more reproducible patient-centered data to aid clinical decision-making and potentially improve quality of care and health status. Poorer health status in the SAQ domains of physical limitation, angina frequency, and quality of life is also associated with graded increases in risk for mortality and rehospitalization for acute coronary syndrome ( Fig. 7.2 ).
